Upload folder using huggingface_hub

.gitattributes

@@ -33,3 +33,4 @@ saved_model/**/* filter=lfs diff=lfs merge=lfs -text
 *.zip filter=lfs diff=lfs merge=lfs -text
 *.zst filter=lfs diff=lfs merge=lfs -text
 *tfevents* filter=lfs diff=lfs merge=lfs -text
+tokenizer.json filter=lfs diff=lfs merge=lfs -text

added_tokens.json

@@ -0,0 +1,24 @@
+{
+  "</tool_call>": 151658,
+  "<tool_call>": 151657,
+  "<|box_end|>": 151649,
+  "<|box_start|>": 151648,
+  "<|endoftext|>": 151643,
+  "<|file_sep|>": 151664,
+  "<|fim_middle|>": 151660,
+  "<|fim_pad|>": 151662,
+  "<|fim_prefix|>": 151659,
+  "<|fim_suffix|>": 151661,
+  "<|im_end|>": 151645,
+  "<|im_start|>": 151644,
+  "<|image_pad|>": 151655,
+  "<|object_ref_end|>": 151647,
+  "<|object_ref_start|>": 151646,
+  "<|quad_end|>": 151651,
+  "<|quad_start|>": 151650,
+  "<|repo_name|>": 151663,
+  "<|video_pad|>": 151656,
+  "<|vision_end|>": 151653,
+  "<|vision_pad|>": 151654,
+  "<|vision_start|>": 151652
+}

chat_template.json

@@ -0,0 +1,3 @@
+{
+  "chat_template": "{% set image_placeholder = '<|vision_start|><|image_pad|><|vision_end|>' %}\n{% for message in messages %}\n    {#--- Handle User Messages with Template and Examples ---#}\n    {%- if message['role'] == 'user' and template -%}\n        {% if loop.first and message['role'] != 'system' %}\n            {{- '<|im_start|>system\\nYou are NuExtract, an information extraction tool created by NuMind.<|im_end|>' }}\n        {% endif %}\n    \n        {{- '<|im_start|>' + message['role'] -}}\n        \n        {#--- Template Section ---#}\n        {{ '\\n# Template:' }}\n        {{- '\\n' + template + '\\n' }}\n        \n        {#--- Examples Section (if provided) ---#}\n        {% if examples -%}\n            {{- '# Examples:' }}\n            {% for example in examples %}\n                {{- '## Input:\\n' }}\n                {#--- Handle image examples ---#}\n                {% if example['input'] is mapping and example['input']['type'] == 'image' %}\n                    {{- image_placeholder | trim -}}\n                {% elif example['input'] == '<image>' %}\n                    {{- image_placeholder | trim -}}\n                {% else %}\n                    {{- example['input'] -}}\n                {% endif %}\n                {{- '\\n## Output:\\n' ~ example['output'] }}\n            {% endfor %}\n        {%- endif %}\n        \n        {#--- Context Section: Handle various content types ---#}\n        {{- '# Context:\\n' }}\n        {%- if message['content'] is string -%}\n            {#--- Simple string content ---#}\n            {{- message['content'] | trim -}}\n        {%- elif message['content'] is mapping and message['content']['type'] == 'image' -%}\n            {#--- Single image document ---#}\n            {{- image_placeholder | trim -}}\n        {%- else -%}\n            {#--- List of content items (mixed text/images) ---#}\n            {#--- First, determine what the actual input content is (not ICL images) ---#}\n            {%- set ns = namespace(has_text_input=false, text_content='') -%}\n            \n            {#--- Count content types and identify actual input document ---#}\n            {%- for content in message['content'] -%}\n                {%- if content is mapping and content.get('type') == 'text' -%}\n                    {%- if content.get('text') != '<image>' -%}\n                        {%- set ns.has_text_input = true -%}\n                        {%- set ns.text_content = content['text'] -%}\n                    {%- endif -%}\n                {%- elif content is string -%}\n                    {%- if content != '<image>' -%}\n                        {%- set ns.has_text_input = true -%}\n                        {%- set ns.text_content = content -%}\n                    {%- endif -%}\n                {%- endif -%}\n            {%- endfor -%}\n            \n            {#--- Determine what to output based on actual input type ---#}\n            {%- if ns.has_text_input -%}\n                {#--- Main input is text, so output the text content ---#}\n                {{- ns.text_content | trim -}}\n            {%- else -%}\n                {#--- Main input is image or <image> placeholder ---#}\n                {%- set ns2 = namespace(found_image=false) -%}\n                {%- for content in message['content'] -%}\n                    {%- if content is mapping and content.get('type') == 'image' and not ns2.found_image -%}\n                        {{- image_placeholder | trim -}}\n                        {%- set ns2.found_image = true -%}\n                    {%- elif content is mapping and content.get('type') == 'text' and content.get('text') == '<image>' and not ns2.found_image -%}\n                        {{- image_placeholder | trim -}}\n                        {%- set ns2.found_image = true -%}\n                    {%- elif content is string and content == '<image>' and not ns2.found_image -%}\n                        {{- image_placeholder | trim -}}\n                        {%- set ns2.found_image = true -%}\n                    {%- endif -%}\n                {%- endfor -%}\n            {%- endif -%}\n        {%- endif -%}\n        {{- '<|im_end|>\\n'}}\n        \n    {#--- Handle All Other Messages (Assistant, System, etc.) ---#}\n    {% else %}\n        {% if loop.first and message['role'] != 'system' %}\n            {{- '<|im_start|>system\\nYou are a helpful assistant.<|im_end|>' }}\n        {% endif %}\n        \n        {{- '<|im_start|>' + message['role'] + '\\n' }}\n        \n        {#--- Same content handling logic as above but without template/examples ---#}\n        {%- if message['content'] is string -%}\n            {{- message['content'] | trim }}\n        {%- elif message['content'] is mapping and message['content']['type'] == 'image' -%}\n            {{- image_placeholder | trim }}\n        {%- else -%}\n            {%- set ns = namespace(found=false) -%}\n            {%- for content in message['content'] -%}\n                {%- if content is string -%}\n                    {{- content | trim -}}\n                {%- elif content is mapping and content.get('type') == 'text' -%}\n                    {{- content['text'] | trim -}}\n                {%- elif content is mapping and content.get('type') == 'image' -%}\n                    {# Skip adding image placeholder - it's already in the text #}\n                {%- endif -%}\n            {%- endfor -%}\n        {%- endif -%}\n        {{- '<|im_end|>'}}\n    {% endif %}\n{% endfor -%}\n{#--- Add Generation Prompt if Requested ---#}\n{%- if add_generation_prompt %}\n    {{- '<|im_start|>assistant' }}\n{% endif -%}"
+}

config.json

@@ -0,0 +1,86 @@
+{
+  "architectures": [
+    "Qwen2_5_VLForConditionalGeneration"
+  ],
+  "attention_dropout": 0.0,
+  "bos_token_id": 151643,
+  "eos_token_id": 151645,
+  "hidden_act": "silu",
+  "hidden_size": 2048,
+  "image_token_id": 151655,
+  "initializer_range": 0.02,
+  "intermediate_size": 11008,
+  "max_position_embeddings": 128000,
+  "max_window_layers": 70,
+  "model_type": "qwen2_5_vl",
+  "num_attention_heads": 16,
+  "num_hidden_layers": 36,
+  "num_key_value_heads": 2,
+  "quantization_config": {
+    "bits": 4,
+    "checkpoint_format": "gptq",
+    "desc_act": true,
+    "group_size": 128,
+    "lm_head": false,
+    "meta": {
+      "damp_auto_increment": 0.0025,
+      "damp_percent": 0.01,
+      "mse": 0.0,
+      "quantizer": [
+        "gptqmodel:2.2.0"
+      ],
+      "static_groups": false,
+      "true_sequential": true,
+      "uri": "https://github.com/modelcloud/gptqmodel"
+    },
+    "pack_dtype": "int32",
+    "quant_method": "gptq",
+    "sym": true
+  },
+  "rms_norm_eps": 1e-06,
+  "rope_scaling": {
+    "mrope_section": [
+      16,
+      24,
+      24
+    ],
+    "rope_type": "default",
+    "type": "default"
+  },
+  "rope_theta": 1000000.0,
+  "sliding_window": 32768,
+  "tie_word_embeddings": true,
+  "torch_dtype": "bfloat16",
+  "transformers_version": "4.51.3",
+  "use_cache": true,
+  "use_sliding_window": false,
+  "video_token_id": 151656,
+  "vision_config": {
+    "depth": 32,
+    "fullatt_block_indexes": [
+      7,
+      15,
+      23,
+      31
+    ],
+    "hidden_act": "silu",
+    "hidden_size": 1280,
+    "in_channels": 3,
+    "in_chans": 3,
+    "intermediate_size": 3420,
+    "model_type": "qwen2_5_vl",
+    "num_heads": 16,
+    "out_hidden_size": 2048,
+    "patch_size": 14,
+    "spatial_merge_size": 2,
+    "spatial_patch_size": 14,
+    "temporal_patch_size": 2,
+    "tokens_per_second": 2,
+    "torch_dtype": "bfloat16",
+    "window_size": 112
+  },
+  "vision_end_token_id": 151653,
+  "vision_start_token_id": 151652,
+  "vision_token_id": 151654,
+  "vocab_size": 151936
+}

generation_config.json

@@ -0,0 +1,13 @@
+{
+  "attn_implementation": "flash_attention_2",
+  "bos_token_id": 151643,
+  "do_sample": true,
+  "eos_token_id": [
+    151645,
+    151643
+  ],
+  "pad_token_id": 151643,
+  "repetition_penalty": 1.05,
+  "temperature": 1e-06,
+  "transformers_version": "4.51.3"
+}

model.safetensors

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:595eec5e2eff5d1eb0a40f112fc904517b01b7226829657941420ad990db13c8
+size 3405167488

preprocessor_config.json

@@ -0,0 +1,29 @@
+{
+  "do_convert_rgb": true,
+  "do_normalize": true,
+  "do_rescale": true,
+  "do_resize": true,
+  "image_mean": [
+    0.48145466,
+    0.4578275,
+    0.40821073
+  ],
+  "image_processor_type": "Qwen2VLImageProcessor",
+  "image_std": [
+    0.26862954,
+    0.26130258,
+    0.27577711
+  ],
+  "max_pixels": 2352000,
+  "merge_size": 2,
+  "min_pixels": 200704,
+  "patch_size": 14,
+  "processor_class": "Qwen2_5_VLProcessor",
+  "resample": 3,
+  "rescale_factor": 0.00392156862745098,
+  "size": {
+    "longest_edge": 2352000,
+    "shortest_edge": 200704
+  },
+  "temporal_patch_size": 2
+}

quant_log.csv

@@ -0,0 +1,253 @@
+layer,module,loss,samples,damp,time
+0,self_attn.k_proj,0.47639775,0.01000,0.535
+0,self_attn.v_proj,0.07883583,0.01000,0.357
+0,self_attn.q_proj,2.47703338,0.01000,0.351
+0,self_attn.o_proj,0.11506961,0.01000,0.349
+0,mlp.up_proj,1.98168492,0.01000,0.359
+0,mlp.gate_proj,2.49659777,0.01000,0.361
+0,mlp.down_proj,0.48027515,0.01000,2.015
+1,self_attn.k_proj,0.12130460,0.01000,0.351
+1,self_attn.v_proj,0.02214950,0.01000,0.346
+1,self_attn.q_proj,0.43165040,0.01000,0.345
+1,self_attn.o_proj,0.03740492,0.01000,0.342
+1,mlp.up_proj,45.12257767,0.01000,0.349
+1,mlp.gate_proj,54.92391968,0.01000,0.354
+1,mlp.down_proj,0.07160743,0.01000,1.997
+2,self_attn.k_proj,0.14272314,0.01000,0.353
+2,self_attn.v_proj,0.03075436,0.01000,0.345
+2,self_attn.q_proj,0.62909025,0.01000,0.352
+2,self_attn.o_proj,0.03786607,0.01000,0.354
+2,mlp.up_proj,19.92619324,0.01000,0.354
+2,mlp.gate_proj,16.12575531,0.01000,0.354
+2,mlp.down_proj,1.23383629,0.01000,2.005
+3,self_attn.k_proj,0.55605268,0.01000,0.351
+3,self_attn.v_proj,0.12625939,0.01000,0.344
+3,self_attn.q_proj,2.61799288,0.01000,0.350
+3,self_attn.o_proj,0.07996529,0.01000,0.353
+3,mlp.up_proj,27.56571579,0.01000,0.360
+3,mlp.gate_proj,26.00288391,0.01000,0.355
+3,mlp.down_proj,0.45218372,0.01000,2.007
+4,self_attn.k_proj,0.42093384,0.01000,0.347
+4,self_attn.v_proj,0.12365282,0.01000,0.347
+4,self_attn.q_proj,1.98233652,0.01000,0.351
+4,self_attn.o_proj,0.18781564,0.01000,0.355
+4,mlp.up_proj,22.99368286,0.01000,0.353
+4,mlp.gate_proj,26.84276772,0.01000,0.348
+4,mlp.down_proj,0.69326061,0.01000,2.001
+5,self_attn.k_proj,0.64253706,0.01000,0.348
+5,self_attn.v_proj,0.18757863,0.01000,0.349
+5,self_attn.q_proj,3.52685070,0.01000,0.358
+5,self_attn.o_proj,0.20589207,0.01000,0.367
+5,mlp.up_proj,36.02760315,0.01000,0.355
+5,mlp.gate_proj,44.12892914,0.01000,0.360
+5,mlp.down_proj,1.28119040,0.01000,2.020
+6,self_attn.k_proj,0.54317248,0.01000,0.350
+6,self_attn.v_proj,0.30935797,0.01000,0.345
+6,self_attn.q_proj,2.96549559,0.01000,0.342
+6,self_attn.o_proj,0.42011255,0.01000,0.356
+6,mlp.up_proj,49.88668823,0.01000,0.357
+6,mlp.gate_proj,61.06584167,0.01000,0.356
+6,mlp.down_proj,1.47490156,0.01000,2.038
+7,self_attn.k_proj,0.76744461,0.01000,0.358
+7,self_attn.v_proj,0.37570187,0.01000,0.366
+7,self_attn.q_proj,4.38668633,0.01000,0.355
+7,self_attn.o_proj,0.26040763,0.01000,0.343
+7,mlp.up_proj,54.43235016,0.01000,0.341
+7,mlp.gate_proj,66.96466064,0.01000,0.345
+7,mlp.down_proj,2.06351209,0.01000,2.019
+8,self_attn.k_proj,0.71847218,0.01000,0.352
+8,self_attn.v_proj,0.46121377,0.01000,0.347
+8,self_attn.q_proj,4.52393961,0.01000,0.361
+8,self_attn.o_proj,0.26604640,0.01000,0.362
+8,mlp.up_proj,57.88217926,0.01000,0.356
+8,mlp.gate_proj,82.81091309,0.01000,0.361
+8,mlp.down_proj,2.05172133,0.01000,2.000
+9,self_attn.k_proj,0.75053275,0.01000,0.355
+9,self_attn.v_proj,0.38040811,0.01000,0.363
+9,self_attn.q_proj,4.08993626,0.01000,0.361
+9,self_attn.o_proj,0.31248963,0.01000,0.354
+9,mlp.up_proj,50.55811310,0.01000,0.367
+9,mlp.gate_proj,76.47357178,0.01000,0.361
+9,mlp.down_proj,2.13050270,0.01000,2.065
+10,self_attn.k_proj,0.84486675,0.01000,0.353
+10,self_attn.v_proj,0.46178406,0.01000,0.352
+10,self_attn.q_proj,5.40872574,0.01000,0.354
+10,self_attn.o_proj,0.37074614,0.01000,0.353
+10,mlp.up_proj,36.80422211,0.01000,0.363
+10,mlp.gate_proj,56.75604248,0.01000,0.360
+10,mlp.down_proj,2.47812700,0.01000,2.026
+11,self_attn.k_proj,0.68764448,0.01000,0.353
+11,self_attn.v_proj,0.43616503,0.01000,0.358
+11,self_attn.q_proj,3.98019719,0.01000,0.354
+11,self_attn.o_proj,0.44827569,0.01000,0.358
+11,mlp.up_proj,21.22748375,0.01000,0.360
+11,mlp.gate_proj,29.39674377,0.01000,0.366
+11,mlp.down_proj,2.60092998,0.01000,2.000
+12,self_attn.k_proj,0.67705953,0.01000,0.348
+12,self_attn.v_proj,0.39055049,0.01000,0.343
+12,self_attn.q_proj,4.22260094,0.01000,0.333
+12,self_attn.o_proj,0.43465185,0.01000,0.348
+12,mlp.up_proj,22.20967484,0.01000,0.356
+12,mlp.gate_proj,31.79467392,0.01000,0.363
+12,mlp.down_proj,2.27273822,0.01000,2.028
+13,self_attn.k_proj,0.84612310,0.01000,0.348
+13,self_attn.v_proj,0.27276921,0.01000,0.353
+13,self_attn.q_proj,4.29369831,0.01000,0.346
+13,self_attn.o_proj,0.27510852,0.01000,0.351
+13,mlp.up_proj,15.75501060,0.01000,0.354
+13,mlp.gate_proj,16.58034897,0.01000,0.355
+13,mlp.down_proj,1.96245372,0.01000,2.117
+14,self_attn.k_proj,0.60780370,0.01000,0.339
+14,self_attn.v_proj,0.27480417,0.01000,0.338
+14,self_attn.q_proj,3.46784306,0.01000,0.341
+14,self_attn.o_proj,0.55041981,0.01000,0.340
+14,mlp.up_proj,14.70653439,0.01000,0.358
+14,mlp.gate_proj,15.58274174,0.01000,0.359
+14,mlp.down_proj,1.81650376,0.01000,1.980
+15,self_attn.k_proj,0.57443047,0.01000,0.350
+15,self_attn.v_proj,0.27507365,0.01000,0.348
+15,self_attn.q_proj,3.37030554,0.01000,0.347
+15,self_attn.o_proj,0.65812862,0.01000,0.353
+15,mlp.up_proj,13.08578873,0.01000,0.360
+15,mlp.gate_proj,12.57400513,0.01000,0.369
+15,mlp.down_proj,1.66309798,0.01000,2.048
+16,self_attn.k_proj,0.62445736,0.01000,0.358
+16,self_attn.v_proj,0.29849333,0.01000,0.345
+16,self_attn.q_proj,3.50871587,0.01000,0.344
+16,self_attn.o_proj,0.56370842,0.01000,0.355
+16,mlp.up_proj,12.91068935,0.01000,0.344
+16,mlp.gate_proj,13.37463760,0.01000,0.353
+16,mlp.down_proj,1.46269417,0.01000,2.004
+17,self_attn.k_proj,0.99160087,0.01000,0.348
+17,self_attn.v_proj,0.36263356,0.01000,0.336
+17,self_attn.q_proj,5.45891762,0.01000,0.345
+17,self_attn.o_proj,0.65637952,0.01000,0.364
+17,mlp.up_proj,11.73399544,0.01000,0.357
+17,mlp.gate_proj,11.55345917,0.01000,0.362
+17,mlp.down_proj,1.54765630,0.01000,1.972
+18,self_attn.k_proj,0.54346764,0.01000,0.340
+18,self_attn.v_proj,0.30677640,0.01000,0.342
+18,self_attn.q_proj,3.33190250,0.01000,0.357
+18,self_attn.o_proj,0.65404612,0.01000,0.358
+18,mlp.up_proj,11.26241779,0.01000,0.358
+18,mlp.gate_proj,11.59079933,0.01000,0.353
+18,mlp.down_proj,1.53053451,0.01000,2.019
+19,self_attn.k_proj,0.68312550,0.01000,0.358
+19,self_attn.v_proj,0.32015702,0.01000,0.340
+19,self_attn.q_proj,3.92224121,0.01000,0.348
+19,self_attn.o_proj,0.92777669,0.01000,0.359
+19,mlp.up_proj,11.23299217,0.01000,0.354
+19,mlp.gate_proj,10.77910995,0.01000,0.354
+19,mlp.down_proj,1.54967749,0.01000,1.989
+20,self_attn.k_proj,0.73232865,0.01000,0.342
+20,self_attn.v_proj,0.62961507,0.01000,0.342
+20,self_attn.q_proj,5.84591770,0.01000,0.344
+20,self_attn.o_proj,1.16900814,0.01000,0.346
+20,mlp.up_proj,12.03665638,0.01000,0.366
+20,mlp.gate_proj,11.68501949,0.01000,0.360
+20,mlp.down_proj,1.81549788,0.01000,1.986
+21,self_attn.k_proj,0.73328865,0.01000,0.343
+21,self_attn.v_proj,0.43035138,0.01000,0.338
+21,self_attn.q_proj,4.56182814,0.01000,0.338
+21,self_attn.o_proj,0.83677453,0.01000,0.366
+21,mlp.up_proj,12.35815144,0.01000,0.363
+21,mlp.gate_proj,13.01117897,0.01000,0.363
+21,mlp.down_proj,1.56445312,0.01000,2.003
+22,self_attn.k_proj,0.84423298,0.01000,0.352
+22,self_attn.v_proj,0.63312274,0.01000,0.345
+22,self_attn.q_proj,5.30216503,0.01000,0.354
+22,self_attn.o_proj,0.88580573,0.01000,0.349
+22,mlp.up_proj,12.89251041,0.01000,0.343
+22,mlp.gate_proj,13.00122261,0.01000,0.335
+22,mlp.down_proj,1.64435768,0.01000,1.954
+23,self_attn.k_proj,0.76918477,0.01000,0.352
+23,self_attn.v_proj,0.34251970,0.01000,0.343
+23,self_attn.q_proj,5.00206470,0.01000,0.341
+23,self_attn.o_proj,0.62123752,0.01000,0.355
+23,mlp.up_proj,13.00738239,0.01000,0.365
+23,mlp.gate_proj,12.38641357,0.01000,0.369
+23,mlp.down_proj,1.91341496,0.01000,2.019
+24,self_attn.k_proj,0.58367682,0.01000,0.361
+24,self_attn.v_proj,0.39665616,0.01000,0.347
+24,self_attn.q_proj,3.92485666,0.01000,0.357
+24,self_attn.o_proj,0.69422036,0.01000,0.349
+24,mlp.up_proj,12.58154488,0.01000,0.355
+24,mlp.gate_proj,12.13731766,0.01000,0.354
+24,mlp.down_proj,1.89018548,0.01000,1.997
+25,self_attn.k_proj,0.55985552,0.01000,0.352
+25,self_attn.v_proj,0.61911839,0.01000,0.360
+25,self_attn.q_proj,5.21534395,0.01000,0.348
+25,self_attn.o_proj,0.71602559,0.01000,0.352
+25,mlp.up_proj,14.45590019,0.01000,0.361
+25,mlp.gate_proj,14.24532509,0.01000,0.354
+25,mlp.down_proj,2.52655506,0.01000,2.048
+26,self_attn.k_proj,0.62120366,0.01000,0.356
+26,self_attn.v_proj,0.67329204,0.01000,0.356
+26,self_attn.q_proj,4.39216709,0.01000,0.354
+26,self_attn.o_proj,0.85128844,0.01000,0.360
+26,mlp.up_proj,15.34780407,0.01000,0.365
+26,mlp.gate_proj,14.43859768,0.01000,0.362
+26,mlp.down_proj,3.05534267,0.01000,2.057
+27,self_attn.k_proj,0.60259527,0.01000,0.363
+27,self_attn.v_proj,0.83670938,0.01000,0.359
+27,self_attn.q_proj,6.72910786,0.01000,0.369
+27,self_attn.o_proj,1.40344274,0.01000,0.368
+27,mlp.up_proj,16.02120972,0.01000,0.372
+27,mlp.gate_proj,15.74603081,0.01000,0.363
+27,mlp.down_proj,3.56427813,0.01000,2.087
+28,self_attn.k_proj,0.61032331,0.01000,0.363
+28,self_attn.v_proj,0.80786330,0.01000,0.345
+28,self_attn.q_proj,4.98351574,0.01000,0.349
+28,self_attn.o_proj,1.22365725,0.01000,0.349
+28,mlp.up_proj,17.86008453,0.01000,0.367
+28,mlp.gate_proj,17.92751694,0.01000,0.360
+28,mlp.down_proj,3.70085478,0.01000,1.981
+29,self_attn.k_proj,0.64153433,0.01000,0.345
+29,self_attn.v_proj,0.73555517,0.01000,0.340
+29,self_attn.q_proj,4.90493393,0.01000,0.341
+29,self_attn.o_proj,0.62442762,0.01000,0.351
+29,mlp.up_proj,20.76404190,0.01000,0.343
+29,mlp.gate_proj,20.62790489,0.01000,0.341
+29,mlp.down_proj,4.93725204,0.01000,2.021
+30,self_attn.k_proj,0.69649112,0.01000,0.351
+30,self_attn.v_proj,1.84374022,0.01000,0.353
+30,self_attn.q_proj,5.63703060,0.01000,0.344
+30,self_attn.o_proj,1.24101484,0.01000,0.351
+30,mlp.up_proj,26.59041214,0.01000,0.361
+30,mlp.gate_proj,25.44146729,0.01000,0.357
+30,mlp.down_proj,13.09652138,0.01000,1.973
+31,self_attn.k_proj,0.88555670,0.01000,0.351
+31,self_attn.v_proj,2.41409063,0.01000,0.342
+31,self_attn.q_proj,6.80245066,0.01000,0.353
+31,self_attn.o_proj,2.95441389,0.01000,0.354
+31,mlp.up_proj,32.29392624,0.01000,0.357
+31,mlp.gate_proj,29.32855034,0.01000,0.357
+31,mlp.down_proj,14.80035019,0.01000,1.949
+32,self_attn.k_proj,1.36434364,0.01000,0.348
+32,self_attn.v_proj,9.14617348,0.01000,0.342
+32,self_attn.q_proj,10.58930969,0.01000,0.353
+32,self_attn.o_proj,3.90258408,0.01000,0.349
+32,mlp.up_proj,36.99082565,0.01000,0.361
+32,mlp.gate_proj,33.20306396,0.01000,0.358
+32,mlp.down_proj,16.74304962,0.01000,2.023
+33,self_attn.k_proj,1.35363579,0.01000,0.355
+33,self_attn.v_proj,15.65682030,0.01000,0.349
+33,self_attn.q_proj,13.18994522,0.01000,0.359
+33,self_attn.o_proj,5.70710468,0.01000,0.354
+33,mlp.up_proj,39.76282883,0.01000,0.354
+33,mlp.gate_proj,33.09777832,0.01000,0.358
+33,mlp.down_proj,38.75576019,0.01000,1.999
+34,self_attn.k_proj,0.69316441,0.01000,0.348
+34,self_attn.v_proj,1.91508794,0.01000,0.338
+34,self_attn.q_proj,6.49875736,0.01000,0.347
+34,self_attn.o_proj,2.32831573,0.01000,0.359
+34,mlp.up_proj,46.86276627,0.01000,0.342
+34,mlp.gate_proj,40.59505844,0.01000,0.344
+34,mlp.down_proj,30.63180161,0.01000,2.010
+35,self_attn.k_proj,0.78077435,0.01000,0.348
+35,self_attn.v_proj,1.97707105,0.01000,0.352
+35,self_attn.q_proj,6.55226660,0.01000,0.353
+35,self_attn.o_proj,1.74587154,0.01000,0.356
+35,mlp.up_proj,64.49967194,0.01000,0.353
+35,mlp.gate_proj,59.11815643,0.01000,0.356
+35,mlp.down_proj,82.74237823,0.01000,1.973

quantize_config.json

@@ -0,0 +1,21 @@
+{
+  "bits": 4,
+  "group_size": 128,
+  "desc_act": true,
+  "sym": true,
+  "lm_head": false,
+  "quant_method": "gptq",
+  "checkpoint_format": "gptq",
+  "pack_dtype": "int32",
+  "meta": {
+    "quantizer": [
+      "gptqmodel:2.2.0"
+    ],
+    "uri": "https://github.com/modelcloud/gptqmodel",
+    "damp_percent": 0.01,
+    "damp_auto_increment": 0.0025,
+    "static_groups": false,
+    "true_sequential": true,
+    "mse": 0.0
+  }
+}

special_tokens_map.json

@@ -0,0 +1,31 @@
+{
+  "additional_special_tokens": [
+    "<|im_start|>",
+    "<|im_end|>",
+    "<|object_ref_start|>",
+    "<|object_ref_end|>",
+    "<|box_start|>",
+    "<|box_end|>",
+    "<|quad_start|>",
+    "<|quad_end|>",
+    "<|vision_start|>",
+    "<|vision_end|>",
+    "<|vision_pad|>",
+    "<|image_pad|>",
+    "<|video_pad|>"
+  ],
+  "eos_token": {
+    "content": "<|im_end|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  },
+  "pad_token": {
+    "content": "<|endoftext|>",
+    "lstrip": false,
+    "normalized": false,
+    "rstrip": false,
+    "single_word": false
+  }
+}

tokenizer.json

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:5eee858c5123a4279c3e1f7b81247343f356ac767940b2692a928ad929543214
+size 11422063

tokenizer_config.json

@@ -0,0 +1,212 @@
+{
+  "add_bos_token": false,
+  "add_prefix_space": false,
+  "added_tokens_decoder": {
+    "151643": {
+      "content": "<|endoftext|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151644": {
+      "content": "<|im_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151645": {
+      "content": "<|im_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151646": {
+      "content": "<|object_ref_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151647": {
+      "content": "<|object_ref_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151648": {
+      "content": "<|box_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151649": {
+      "content": "<|box_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151650": {
+      "content": "<|quad_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151651": {
+      "content": "<|quad_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151652": {
+      "content": "<|vision_start|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151653": {
+      "content": "<|vision_end|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151654": {
+      "content": "<|vision_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151655": {
+      "content": "<|image_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151656": {
+      "content": "<|video_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": true
+    },
+    "151657": {
+      "content": "<tool_call>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151658": {
+      "content": "</tool_call>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151659": {
+      "content": "<|fim_prefix|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151660": {
+      "content": "<|fim_middle|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151661": {
+      "content": "<|fim_suffix|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151662": {
+      "content": "<|fim_pad|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151663": {
+      "content": "<|repo_name|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    },
+    "151664": {
+      "content": "<|file_sep|>",
+      "lstrip": false,
+      "normalized": false,
+      "rstrip": false,
+      "single_word": false,
+      "special": false
+    }
+  },
+  "additional_special_tokens": [
+    "<|im_start|>",
+    "<|im_end|>",
+    "<|object_ref_start|>",
+    "<|object_ref_end|>",
+    "<|box_start|>",
+    "<|box_end|>",
+    "<|quad_start|>",
+    "<|quad_end|>",
+    "<|vision_start|>",
+    "<|vision_end|>",
+    "<|vision_pad|>",
+    "<|image_pad|>",
+    "<|video_pad|>"
+  ],
+  "bos_token": null,
+  "chat_template": "{% set image_placeholder = '<|vision_start|><|image_pad|><|vision_end|>' %}\n{% for message in messages %}\n    {#--- Handle User Messages with Template and Examples ---#}\n    {%- if message['role'] == 'user' and template -%}\n        {% if loop.first and message['role'] != 'system' %}\n            {{- '<|im_start|>system\\nYou are NuExtract, an information extraction tool created by NuMind.<|im_end|>' }}\n        {% endif %}\n    \n        {{- '<|im_start|>' + message['role'] -}}\n        \n        {#--- Template Section ---#}\n        {{ '\\n# Template:' }}\n        {{- '\\n' + template + '\\n' }}\n        \n        {#--- Examples Section (if provided) ---#}\n        {% if examples -%}\n            {{- '# Examples:' }}\n            {% for example in examples %}\n                {{- '## Input:\\n' }}\n                {#--- Handle image examples ---#}\n                {% if example['input'] is mapping and example['input']['type'] == 'image' %}\n                    {{- image_placeholder | trim -}}\n                {% elif example['input'] == '<image>' %}\n                    {{- image_placeholder | trim -}}\n                {% else %}\n                    {{- example['input'] -}}\n                {% endif %}\n                {{- '\\n## Output:\\n' ~ example['output'] }}\n            {% endfor %}\n        {%- endif %}\n        \n        {#--- Context Section: Handle various content types ---#}\n        {{- '# Context:\\n' }}\n        {%- if message['content'] is string -%}\n            {#--- Simple string content ---#}\n            {{- message['content'] | trim -}}\n        {%- elif message['content'] is mapping and message['content']['type'] == 'image' -%}\n            {#--- Single image document ---#}\n            {{- image_placeholder | trim -}}\n        {%- else -%}\n            {#--- List of content items (mixed text/images) ---#}\n            {#--- First, determine what the actual input content is (not ICL images) ---#}\n            {%- set ns = namespace(has_text_input=false, text_content='') -%}\n            \n            {#--- Count content types and identify actual input document ---#}\n            {%- for content in message['content'] -%}\n                {%- if content is mapping and content.get('type') == 'text' -%}\n                    {%- if content.get('text') != '<image>' -%}\n                        {%- set ns.has_text_input = true -%}\n                        {%- set ns.text_content = content['text'] -%}\n                    {%- endif -%}\n                {%- elif content is string -%}\n                    {%- if content != '<image>' -%}\n                        {%- set ns.has_text_input = true -%}\n                        {%- set ns.text_content = content -%}\n                    {%- endif -%}\n                {%- endif -%}\n            {%- endfor -%}\n            \n            {#--- Determine what to output based on actual input type ---#}\n            {%- if ns.has_text_input -%}\n                {#--- Main input is text, so output the text content ---#}\n                {{- ns.text_content | trim -}}\n            {%- else -%}\n                {#--- Main input is image or <image> placeholder ---#}\n                {%- set ns2 = namespace(found_image=false) -%}\n                {%- for content in message['content'] -%}\n                    {%- if content is mapping and content.get('type') == 'image' and not ns2.found_image -%}\n                        {{- image_placeholder | trim -}}\n                        {%- set ns2.found_image = true -%}\n                    {%- elif content is mapping and content.get('type') == 'text' and content.get('text') == '<image>' and not ns2.found_image -%}\n                        {{- image_placeholder | trim -}}\n                        {%- set ns2.found_image = true -%}\n                    {%- elif content is string and content == '<image>' and not ns2.found_image -%}\n                        {{- image_placeholder | trim -}}\n                        {%- set ns2.found_image = true -%}\n                    {%- endif -%}\n                {%- endfor -%}\n            {%- endif -%}\n        {%- endif -%}\n        {{- '<|im_end|>\\n'}}\n        \n    {#--- Handle All Other Messages (Assistant, System, etc.) ---#}\n    {% else %}\n        {% if loop.first and message['role'] != 'system' %}\n            {{- '<|im_start|>system\\nYou are a helpful assistant.<|im_end|>' }}\n        {% endif %}\n        \n        {{- '<|im_start|>' + message['role'] + '\\n' }}\n        \n        {#--- Same content handling logic as above but without template/examples ---#}\n        {%- if message['content'] is string -%}\n            {{- message['content'] | trim }}\n        {%- elif message['content'] is mapping and message['content']['type'] == 'image' -%}\n            {{- image_placeholder | trim }}\n        {%- else -%}\n            {%- set ns = namespace(found=false) -%}\n            {%- for content in message['content'] -%}\n                {%- if content is string -%}\n                    {{- content | trim -}}\n                {%- elif content is mapping and content.get('type') == 'text' -%}\n                    {{- content['text'] | trim -}}\n                {%- elif content is mapping and content.get('type') == 'image' -%}\n                    {# Skip adding image placeholder - it's already in the text #}\n                {%- endif -%}\n            {%- endfor -%}\n        {%- endif -%}\n        {{- '<|im_end|>'}}\n    {% endif %}\n{% endfor -%}\n{#--- Add Generation Prompt if Requested ---#}\n{%- if add_generation_prompt %}\n    {{- '<|im_start|>assistant' }}\n{% endif -%}",
+  "clean_up_tokenization_spaces": false,
+  "eos_token": "<|im_end|>",
+  "errors": "replace",
+  "extra_special_tokens": {},
+  "max_pixels": 2352000,
+  "min_pixels": 200704,
+  "model_max_length": 131072,
+  "pad_token": "<|endoftext|>",
+  "padding_side": "left",
+  "processor_class": "Qwen2_5_VLProcessor",
+  "split_special_tokens": false,
+  "tokenizer_class": "Qwen2Tokenizer",
+  "unk_token": null
+}